Correctness-preserving derivation of concurrent garbage collection algorithms

Abstract

Constructing correct concurrent garbage collection algorithms is notoriously hard. Numerous such algorithms have been proposed, implemented, and deployed -and yet the relationship among them in terms of speed and precision is poorly understood, and the validation of one algorithm does not carry over to others.As programs with low latency requirements written in garbagecollected languages become part of society's mission-critical infrastructure, it is imperative that we raise the level of confidence in the correctness of the underlying system, and that we understand the trade-offs inherent in our algorithmic choice.In this paper we present correctness-preserving transformations that can be applied to an initial abstract concurrent garbage collection algorithm which is simpler, more precise, and easier to prove correct than algorithms used in practice -but also more expensive and with less concurrency. We then show how both pre-existing and new algorithms can be synthesized from the abstract algorithm by a series of our transformations. We relate the algorithms formally using a new definition of precision, and informally with respect to overhead and concurrency.This provides many insights about the nature of concurrent collection, allows the direct synthesis of new and useful algorithms, reduces the burden of proof to a single simple algorithm, and lays the groundwork for the automated synthesis of correct concurrent collectors.

Relaxing safely: verified on-the-fly garbage collection for x86-TSO

“…Vechev et al [38] explore correctness-preserving transformations that allow synthesis of a diverse range of more realistic collectors (less expensive, more concurrent), from an apex abstract collector that is simpler, and easier to prove correct. While interesting as a means to enumerating possible designs, this approach does not yield a collector with the fine-grained concurrency and mutator responsiveness of our collector.…”

Bonsai: synthesis-based reasoning for type systems

“…Generation of a counterexample program is related to program synthesis because we can think of the type system as the speci cation: the desired program passes the typechecker and fails in the interpreter. Synthesizers can be roughly divided into three kinds: rewriting [20,23,42], deductive [33,34,51], and those based on searching a space of programs with contraint solvers [7,25,50,51,60]. While the rst two categories derive a program from a speci cation, the last category searches a space of programs, conceptually evaluating each against the speci cation.…”

Adaptive GC-Aware Load Balancing Strategy for High-Assurance Java Distributed Systems

“…Multiple research efforts have focused on improving the performance of the GC: Several works have proposed new parallel [2], [18] and concurrent algorithms [14], [21] that have smaller impacts on the performance of the applications. Another approach has been to develop algorithms that might have predictable GC performance [8].…”